Yimin Dai

National University of Singapore, Nanyang Technological University

Papers

Total Citations

H-Index

About

Yimin Dai is a leading researcher at the intersection of cybersecurity, privacy, and machine learning, with a primary focus on exposing novel side-channel attacks in smart devices. Dai’s most impactful contribution is the development of **LidarPhone**, a groundbreaking acoustic side-channel attack that exploits a robot vacuum cleaner’s lidar sensor to eavesdrop on private conversations. This work, published in 2020 and garnering **83 citations**, demonstrated that lidar—a technology designed for navigation—can be repurposed to detect minute vibrations from sound sources, effectively turning a household appliance into a surveillance tool. Dai further refined this attack in a follow-up paper (11 citations), highlighting the growing threat of non-traditional eavesdropping vectors. More recently, Dai has advanced into **stochastic differential equation networks (SDENets)** for edge computing, addressing stability and computational efficiency in continuous-time neural networks. This work, published in 2025, showcases Dai’s versatility in bridging theoretical machine learning with practical, privacy-critical applications. Recognized for exposing overlooked vulnerabilities in consumer IoT devices, Dai’s research serves as a critical wake-up call for both industry and academia, emphasizing the urgent need for robust privacy safeguards in our increasingly connected world.